Peter Knight's Web Site
Things for me, friends, family... and passers by

decorative bar

Peter Knight's Research

Selected Extracts from Publications


Knight, P.G. (1997)  Quaternary Science Reviews, Vol. 16,  975-993. (Elsevier Science Ltd.)



 In many glaciers and ice sheets there is a basal ice layer (BIL) in which the ice is conditioned primarily by processes operating at the bed. The BIL is chemically and physically distinctive, and is characterised by a component of basally derived sediment. The BIL is: a rheological control on ice-sheet dynamics; an indicator of  subglacial conditions and processes; an agent of subglacial geologic processes; the source of a substantial proportion of glacial sediments; a limit to the downward extension of the climate record from deep ice cores. If debris characteristics of the BIL are preserved in glacial sediments, former glacier conditions can be inferred. 
 Analysis of stable isotopes, gas and cation composition, ice crystallography, debris composition and structural glaciology has elucidated many mechanisms for BIL formation. Common elements can be identified from a range of locations. Key issues are entrainment of ice and debris from the bed, formation of new ice, metamorphism of existing ice in the BIL, and thickening of the BIL by deformation. A distinction can be made between basal ice formed by accretion of material from the bed (“stratified facies”) and basal ice formed by metamorphic processes within the ice close to the bed (“dispersed facies”). 




 Theoretical models have suggested that subglacial water can enter basal ice along crystal boundaries. This has previously been proposed as a mechanism for entraining debris into the base of ice sheets without entraining layers of frozen meltwater. We describe two sets of laboratory experiments designed to establish whether the hypothesis that sediment entrainment into polycrystalline ice could occur along crystal boundaries is mechanically realistic. One set of experiments explores the response of basal sediment to the vertical passage of thermal fronts. A second set explores the response of basal sediment to the imposition of different pressure gradients through the ice. 
 The experiments demonstrate that debris particles of 30-70 microns can be entrained upwards from subglacial sediment into the vein network of polycrystalline ice in response to both thermal and pressure gradients. The calibre of the sediment entrained is controlled by vein diameter, and the disposition of sediment within the ice is controlled by the geometry of the vein network. The entrainment process may be sensitive not only to absolute temperature and pressure values but to the strength of thermal and pressure gradients. At the resolution of the experiments so far carried out, no alteration of the ice is evident other than at crystal boundaries. 

KEYWORDS: subglacial debris entrainment; polycrystalline ice; intercrystal vein network; glacier basal ice. 


Knight, P.G. and Knight D.A. (1994)  Journal of Glaciology 40 (136), 600-601. 


Recent theoretical modelling by Lliboutry (1993) has suggested that standard models of glacier sliding by regelation are flawed. Lliboutry's analysis predicts that melting of ice in response to flow obstruction by bed protruberances will occur in a layer of thickness h(w), through which water will be mobile in the vein capillary network....     Lliboutry has called for field observation of h(w), to constrain his model, but has recognised  problems in identifying h(w) in the field...    We suggest that observations of the dispersed facies of the basal layer, which is accessible at some glacier margins, could provide the modelling constaint that Lliboutry has called for and a test of his theory....  We have carried out a series of experiments in the low-temperature laboratory that demonstrate both debris transport through the vein network of polycrystalline ice and entrainment of debris into the vein network from beneath the ice in response to a pressure gradient....  If Lliboutry's zone h(w) does exist, then the dispersed facies may be a visible consequence of it.   ....estimates of h(w) should be possible on the basis of field observations at ice-margin sites and from deep cores. 

Knight, P.G. (1995) Boreas 24, 11-12. 


The basal ice of many glaciers contains debris structures that reflect subglacial processes. Presented here is an unusually clear photograph of ice and debris in the lowest 2m of the basal layer at the margin of the Greenland ice sheet. The photograph shows ice-debris relationships and deformation structures that reflect entrainment processes and flow history.




Knight, P.G. (1994) Geology, Vol. 22, pp.971-974


Two exposures of basal ice in Alaska and Greenland, which have previously provided the basis for contrasting models of basal-ice development, are in fact directly comparable. Reinterpretation of previous field descriptions, combined with new structural and sedimentological data from West Greenland, facilitates a simple unification of previous models of basal-ice development and a common terminology for the field description of basal-ice sequences. The stratigraphy of the basal-ice sequence at any site indicates the subglacial conditions and processes operating up-glacier of that site. Presented here are interpretations of the major sequence types.




Knight, P.G. (1989) Journal of Glaciology Vol.35 No.120, pp. 214-216.


This paper tests and falsifies the theory that the development of thick sequences of vertically stacked clean and debris-laden layers at the  margin of the Greenland ice sheet can be attributed solely to simple freezing-on of material at the bed. Isotopic analysis in  oxygen-18 and  deuterium of ice from the ice-sheet margin near Sondre Stromfjord, indicates that the debris-rich and debris-poor elements of the basal sequence have different origins. While the debris bands display isotopic fractionation consistent with a freezing origin, the intercalated clean ice layers do not. The clean ice layers have isotopic values indistinguishable from debris-bearing ice immediately above the debris-band sequence and from unaltered glacier ice, and are entrained by a different process from the debris bands. Debris may be entrained by freezing at the bed, but the development of a vertically stacked sequence of debris bands must be attributed to some other mechanism.



Knight, P.G. (1987) In: Waddington, E.D. and Walder, J.S. "The Physical basis of Ice Sheet Modelling." International Association of Hydrological Sciences Publication 170, pp.359-366. 

Abstract Basal ice sequences at the margin of the Greenland ice sheet reflect both marginal and interior basal processes, and have important boundary condition implications for modelling. Analysis of stable isotopes (deuterium, oxygen-18) and debris from the margin indicates two separate zones of debris entrainment beneath the ice; a zone of bulk adfreezing near to the margin, and a zone of warm-bed regelation in the interior. Surface strain measurements indicate two predominant marginal  flow regimes. Widespread compressive flow is accentuated during the winter by basal adfreezing, with major folding extending up to 1km from the margin, and intense deformation in the extreme marginal zone giving rise to thick stacking of both marginally and interior derived  basal layers. By contrast, where rapid streaming flow reaches the margin, extending flow predominates, and bottom melting removes the lower layers of the interior derived regelation ice. 


Knight, P.G. (1998) Progress in Physical Geography 22 (3), 407-411

Section Headings

              Co-operative glaciologists 
              The Quaternary and geomorphology 
              Extra-terrestrial glaciers 
              Siege glaciology 
              The future 
Selections from the text:

Although the study of glaciers draws expertise from a wide range of disciplines, the number of researchers who would explicitly call themselves glaciologists is relatively small. The International Glaciological Society, which is the principal professional organisation for glaciologists,  has a membership of well under 1000.   Hughes (1985, p.39) suggested that “glaciology is a small profession”, and in numerical terms he is right. In practical terms, however, the boundaries of  glaciology are more difficult to delimit. It is a discipline with tentacles that  intertwine with those of many others, and the cultivation of these links has been a prominent feature of recent developments in the broad field of glacier-related studies. A number of new initiatives and publications have seen collaboration between glaciologists and colleagues from neighbouring disciplines, and some of these have involved the continuing expansion of the realm of glaciology beyond terrestrial confines.  This review will focus on some areas of  recent and future cross-discipline co-operation and dialogue. 

 A major part of modern research effort is dedicated to modelling glacier behaviour. Models are quintessentially  collaborative constructions, derived from the integration of diverse elements.  A great ice-sheet model would be like a symphony, drawing together into a unified picture the disparate themes of the glacier story. Thermal regime, hydrology, substrate rheology, ice deformation, mass balance, geomorphology, and glacier dynamics combine transparently into a single system. Creating an ice-sheet model is like translating an epic natural poem into something understandable, and usable,  at a human scale. Modelling either of whole ice sheets or of components of the glacier system aims to facilitate predictions and reconstructions of glacier behaviour. Adequate modelling for these purposes requires some understanding of the processes and parameters that control glacier behaviour. Developments in modelling procedures need to progress hand in hand with observational glaciological data. Modelling, like remote sensing, needs ground-truth data. Models, like engines, require fuel, and the fuel of a good model is good base-line data. 

 A striking feature of modern glaciology is the tendency for groups of collaborators to focus their attentions over prolonged periods at individual sites. A part of this trend in modern glaciological research programmes has been the ascendancy of what can be called “Siege Glaciology”.  This style of research involves comprehensive, long-term, data acquisition on a range of topics at a specific site. Base camps are established, teams of researchers are co-ordinated over periods of several years to handle different aspects of the multi-analytical scientific and technical programmes, and long-term funding and management strategies are employed. In modern glaciology it is a style of research that can be attributed in part to the logistical requirements of collaborative, technology-based field research. Once a logistical and administrative infrastructure for research at a particular site is established, the use of the site becomes self-reinforcing. Once a glacier has been well studied, the data that has been collected there serves as a framework to encourage further study. Furthermore, there is a preference on the part of many glaciologists for multi-parameter research. Because so many aspects of the glacier system are closely interlinked, efforts to understand one part of the system are increasingly deemed to require data from other parts of the system. No self-respecting program of study into the mechanisms of basal motion could proceed without an array of boreholes to monitor basal water pressure, or ploughmeters to test the substrate, or equipment to monitor glaci-seismic activity. Hence major research programs grow up to involve large groups of researchers around a relatively small number of glaciers. There are many examples of this. Reading through the literature of recent years one quickly becomes familiar with a mere handful of glaciers: Storglaciären, Haut Glacier d’Arolla, Variegated Glacier, Jakobshavn Isbrae.  In the ice-sheet literature one quickly gets to know the Siple Coast ice streams, the GRIP and GISP2 cores, Vostok, and a few other sites. Each of these, and a small number of others, have huge lists of publications attaching to them, while a huge number of other glaciers or ice-sheet locations are more or less unmentioned in the literature. Research at Haut Glacier d’Arolla, Switzerland, for example, has been generating research publications at an average rate of about one per month throughout the present decade. Siege programmes now constitute a substantial proportion of the research being carried out on glaciers. 

Glaciology has yet to take full advantage of many possible avenues of interdisciplinary collaboration. There have been substantial contributions to glaciology from physicists, chemists and mathematicians, and there has been some interdisciplinary dialogue in fields such as fracture mechanics and creep (e.g. Lliboutry, 1987), but much potentially valuable interdisciplinary collaboration in the fields of material sciences has yet to be explored. Most of the glaciologists who study glaciers in the field have little contact with ceramicists, metallurgists or others who study equivalent phenomena in other materials. Cross-boundary collaborations on physical and chemical phenomena and in mathematical or technical procedures  have not widely translated to the scale of whole glaciers or whole physical systems. The sort of collaboration that is possible has been demonstrated recently by collaboration between glaciologists, hydrologists and chemists in glacier hydrology and in ice-core analysis through the application of chemical principals to the study of glaciers. There are further clues to glacier behaviour hidden in the physical characteristics of the ice that are not yet being adequately interpreted. Crystallographers and structural glaciologists will in future throw as much new light on glacier behaviour as chemists have done in recent years. 


Glacier advance, ice-marginal lakes and routing of meltwater and sediment: Russell Glacier, Greenland.

Knight, P.G., Waller, R.I., Patterson, C.J., Jones, A.P. and Robinson, Z.P. (2000) 

JOURNAL OF GLACIOLOGY 46 (154), 423-426. 


The ice-sheet margin at Russell Glacier, West Greenland, advanced ~7 m/a between 1968 and 1999. As the ice advanced over moraine ridges, small changes in position caused major changes in the routing of proglacial water and sediemnt. These included changes in distribution of ice-marginal lakes, in the periodic drainage of ice-marginal lakes, in the routing and sediment content of water draining into the proglacial zone, and in the release of sediment from the moraines by erosion and mass movements. Proglacial hydrology and sediment flux appear to be controlled not simply by glacier mass balance, but by evolving ice-marginal geomorphology, which must be accounted for in palaeoenvironmental interpretation of proglacial sediments.

Preservation of basal-ice sediment texture in ice-sheet moraines

Peter G. Knight, Carrie J. Patterson, Richard I. Waller, Alison P. Jones,
Zoe P. Robinson



Ice-sheet moraines near Kangerlussuaq in west Greenland inherit distinctive particle-size distributions from basal ice, although debris structures from the basal ice are commonly destroyed by deposition and resedimentation processes. The abundance of clay and silt in the dispersed facies  basal ice at the ice-sheet margin is clearly reflected in the sedimentology of the ice-sheet moraine. Geographical variations in the texture or grain size of moraine sediments may thus reflect geographical variations in basal ice. This offers a new approach to reconstructing the basal-ice characteristics, and hence the thermal and dynamic properties, of former ice sheets. 

Changes in sediment routing as a consequence of ice-sheet advance, Russell Glacier, Greenland 

Knight, P. G.,  Patterson, C. J., and Waller, R. I. 

Eos Trans. AGU, 82 (47), Fall Meet. Suppl., Abstract (2001)

An advancing ice margin is an extremely complex sedimentary environment. As the position of the ice margin changes through time, debris sources, sediment transfer routes and depositional environments can vary rapidly. The position of an advancing ice-sheet margin with respect to its ancient marginal moraines controls the way sediment is released from the basal ice as well as where and how long it is stored. On the basis of our observations at the edge of the Greenland ice sheet we have identified six stages in the advance of the glacier that have distinctive sedimentary processes. For each stage we describe: (1) debris release, routing, and deposition close to the margin; and (2) processes of erosion, mass movement and deposition that affect the length of time that the sediment is stored in the ice-proximal environment. The progress of the ice across the moraine causes changes in ice-proximal processes that are recorded in the sedimentary record, and we find that the overtopping of moraine ridges has significant consequences for both proximal and distal proglacial environments. Sediment production at this land-based section of the Greenland ice-sheet margin is dominated by debris released through the basal ice layer, which is up to 30 m thick. The debris flux through the basal ice is 21.6 m3m-1yr-1 or approximately 75.6 x 103 kg m-1yr-1. Only negligible amounts are released through englacial, supraglacial or subglacial transfer. Glaciofluvial sediment production is highly localized and long sections of the ice margin receive no sediment from glaciofluvial sources. 

Discharge of debris from ice at the margin of the Greenland ice sheet.

Knight, P.G., Waller, R.I., Patterson, C.J., Jones, A.P. and Robinson, Z.P. (2002)
JOURNAL OF GLACIOLOGY 48 (161), 192-198.

ABSTRACT.  Sediment production at a terrestrial section of the ice-sheet margin in West Greenland is dominated by debris released through the basal ice layer. The debris flux through the basal ice at the margin is estimated to be  12-45 m3 m-1 a-1. This is three orders of magnitude higher than that previously reported for East Antarctica, an order of magnitude hugher than sites reported from Norway, Iceland and Switzerland, but an order of magnitude lower than values previously reported from tidewater glaciers in Alaska and other high-rate environments such as surging glaciers. At our site, only negligible amounts of debris are released through englacial, supraglacial or subglacial sediment transfer. Glaciofluvial sediment production is highly localised, and long sections of the ice-sheet margin receive no sediment from glaciofluvial sources. These findings differ from those of studies at more temperate glacial settings where glaciofluvial routes are dominant and basal ice contributes only a minor percentage of the debris released at the margin. These data on debris flux through the terrestrial margin of an outlet glacier contribute to our limited knowledge of debris production from the Greenland ice sheet.

Knight, P.G. and Knight, D.A. (2004) 
Field observations and laboratory simulations of basal ice formed by 
freezing of supercooled subglacial water 

Invited contribution to AMICS (Antarctic ice-sheet dynamics and climatic change: Modelling and Ice Composition Studies) workshop Dynamic Interaction between the Antarctic Ice Sheet and the Subglacial Environment, Vrije Universiteit Brussel, Brussels, April 2004. Sponsored by the Belgian Federal Science Policy Office (BELSPO). See Powerpoint Slides from presentation here (2MB .pdf file)

It has been suggested that supercooling of meltwater in overdeepened subglacial basins can freeze large quantities of ice and debris to the glacier bed. This process has been invoked to explain high sediment flux from the Laurentide ice sheet, and in a modern glacial context it has been suggested, for example, that all of the basal ice at some Icelandic glaciers can be attributed to subglacial supercooling. To test this hypothesis we compared basal ice facies at supercooling sites in Iceland with ice facies created experimentally by supercooling in the laboratory. We found that specific, distinctive, basal ice facies appear to be created directly by freezing of supercooled water, but that these facies account for a only small proportion of the total basal ice sequence. We are now designing experiments to test whether the remaining basal ice could be created by a multi-stage process with supercooling as the primary entrainment mechanism, or whether mechanisms such as regelation and flow diagenesis, not requiring supercooling, remain tenable. At issue is the question of whether basal ice characteristics in modern glaciers, and sediment flux signatures from former glaciers, can be used as evidence of glaciohydraulic supercooling in the way that has recently been proposed.

Glaciers: art and history, science and uncertainty 
Peter G. Knight 

Interdisciplinary Science Reviews, December 2004, vol. 29, no. 4, pp. 385-393(9) 


Glaciers play a central role in the global environmental system, and their behaviour is intimately linked to changing patterns of the ocean–atmosphere circulation, climate, sea level and landscape. Deep cores retrieved from ice sheets have helped us recognise how climate has changed over hundreds of thousands of years, and, for many parts of the Earth's surface, an understanding of landforms, drainage patterns and surface geology would be impossible without an understanding of glacial processes. Today, the role of glaciers in the media and in the popular imagination is dominated by their role in science as indicators of environmental change. However, it is barely a hundred and fifty years since the significance of glaciers in this context was first appreciated. Before the middle of the nineteenth century glaciers occupied a different niche in popular perception, ruled by their place in the artistic and cultural domain rather than the scientific. Changing technologies for observing and analysing glacial phenomena have impacted on both scientific and cultural perceptions of glaciers, and our understanding is still constrained by technological limitations. Despite progress in remote sensing and analytical techniques, our reconstructions of past glaciations remain tentative, our understanding of modern glacial processes incomplete and our modelling of their future unreliable. In nineteenth century art, glaciers represented romance, mystery and unassailable majesty. In twenty-first century science their position is perhaps similar, but what art calls 'mystery', science calls 'uncertainty'.